File:  [Repository] / ECHO_content / ECHO_language.py
Revision 1.1: download - view: text, annotated - select for diffs - revision graph
Wed Nov 24 15:17:41 2004 UTC (19 years, 5 months ago) by dwinter
Branches: MAIN
CVS tags: HEAD
new modoule with language technologies

"""Methoden fuer Language Technologies"""
def donatus(txt2):
	import xmlrpclib

	server = xmlrpclib.ServerProxy("http://archimedes.fas.harvard.edu/cgi-bin/donatus-rpc")

	txt=txt2.encode('utf-8')
	bin=xmlrpclib.Binary(txt)

	

	ret=server.donatus.analyze(bin)

		
	return ret['morphData'].data


def donatusVariant2Lemma(morphData):
	"""creates hash variant -> morphdata"""
	ret={}
	dom=xml.dom.minidom.parseString(morphData)
	lemmas=dom.getElementsByTagName('lemma')
	for lemma in lemmas:
		variants=lemma.getElementsByTagName('variant')
		for variant in variants:
			atr=variant.getAttribute('form')
			if ret.has_key(atr):
				ret[atr].append=lemma.getAttribute('form')
			else:
				ret[atr]=[lemma.getAttribute('form')]

	return ret

FreeBSD-CVSweb <freebsd-cvsweb@FreeBSD.org>